With the creation of intelligent digital personal assistants, (e.g., SIRI, S Voice, GOOGLE NOW, CORTANA, and HIDI) the use of voice commands to control electronic devices has become extremely popular. SIRI is a registered trademark of Apple Inc. in the United States and other countries. S VOICE is a registered trademark of Samsung Electronics Co. in the United States and other countries. GOOGLE is a registered trademark of Google Inc. in the United States and other countries. CORTANA is a pending trademark of Microsoft in the United States and other countries. Generally, a user interacts with a voice input module, for example embodied in a personal assistant through use of natural language. This style of interface allows a device to receive voice inputs, e.g., voice commands, from the user (e.g., “What is the weather tomorrow,” “Call Dan”) process those requests and perform the user's desired actions by carrying out the task itself or delegating user requests to a desired application.
Because natural language is a major method of communication that people are comfortable with, the ability to use voice commands offers a natural and efficient way to utilize functions of a device's operating system or applications, no matter how simple or complex. However, one of the major issues when utilizing the personal assistants is determining what portion of a user's speech is intended to be received as a voice command. Constantly listening to the user has proven too difficult a task to achieve with a usable level of false positives (i.e., the assistant responding to unrelated speech) and false negatives (i.e., the assistant ignoring user commands). In addition, the personal assistant can be an energy intensive application, thus allowing it to run constantly in the background could have a significant impact on battery life. In order to overcome this issue, most voice controlled assistants today make use of some form of trigger to initiate the voice recognition process. This trigger assumes that any speech directly following the trigger is a command directed to the assistant. Some common triggers are physical button presses (e.g., SIRI activation) and special key phrases to be spoken before any system-directed command (e.g., Okay GOOGLE).